Clin Transl Sci. 2021 Mar 30. doi: 10.1111/cts.13030. Online ahead of print.
Sepsis is a major cause of mortality among hospitalized patients worldwide. Shorter time to administration of broad-spectrum antibiotics is associated with improved outcomes, but early recognition of sepsis remains a major challenge. In a two-center cohort study with prospective sample collection from 1400 adult patients in emergency departments suspected of sepsis, we sought to determine the diagnostic and prognostic capabilities of a machine- learning algorithm based on clinical data and a set of uncommonly measured biomarkers. Specifically, we demonstrate that a machine-learning model developed using this dataset outputs a score with not only diagnostic capability but also prognostic power with respect to hospital length of stay (LOS), thirty-day mortality, and thirty-day inpatient readmission both in our entire testing cohort and various subpopulations. The area under the Receiver Operating Curve (AUROC) for diagnosis of sepsis was 0.83. Predicted risk scores for patients with septic shock were higher compared to patients with sepsis but without shock (p < 0.0001). Scores for patients with infection and organ dysfunction were higher compared to those without either condition (p < 0.0001). Stratification based on predicted scores of the patients into low, medium and high-risk groups showed significant differences in length of stay (p < 0.0001), thirty-day mortality (p < 0.0001), and thirty-day inpatient readmission (p < 0.0001). In conclusion, a machine-learning algorithm based on EMR data and three non-routinely measured biomarkers demonstrated good diagnostic and prognostic capability at the time of initial blood culture.